'The Farm beneath the Sand'-An Archaeological Case Study on Ancient 'Dirt' DNA

Article excerpt


It is generally known among archaeologists that ancient DNA can be obtained from macrofossil remains such as bones and teeth. It is perhaps less recognised that ancient DNA can also be retrieved directly from ancient sediments, ice and faeces, even in the absence of visible macrofossils (commonly referred to as ancient 'dirt' DNA) (e.g. Hofreiter et al. 2003; Willerslev et al. 2003, 2007; Gilbert et al. 2008). For example, it has been shown that ancient DNA deriving from diverse micro-organisms, various plants and vertebrates, including mammoth, horse, bison and musk oxen, can be retrieved directly from small amounts (less than 2g) of sediments both under frozen and non-frozen conditions (Hofreiter et al. 2003; Willerslev et al. 1999, 2003, 2004a & b; Lydolph et al. 2005; Hansen et al. 2006; Haile et al. 2007; Johnson et al. 2007). This has allowed detailed reconstructions of palaeo-ecosystems in Siberia, North America and New Zealand and makes it possible, for the first time, to link past animals and plants in time and space, even in the absence of macrofossil evidence (Willerslev & Cooper 2005). Likewise, DNA obtained directly from 450-800 000 year-old silty ice, stored at the base of the Greenland ice sheet, has revealed the youngest evidence of conifer forest in Greenland (Willerslev et al. 2007) and DNA from ancient faeces was recently used to push back the time of the first peopling of North America by more than 1000 years (Gilbert et al. 2008). Thus, the 'dirt' DNA approach has proved highly significant in improving our understanding of general evolutionary processes.

Although it has been shown that DNA in sediments can account for up to 10 per cent of extractable phases (Trevors 1996; Turner & Newman 2005), it remains partly unclear how DNA from plants and animals may end up in sediments in sufficient quantities for them to be detected by standard molecular techniques, thousands of years after deposition (Willerslev et al. 2004b). Theoretical considerations, coupled with studies on DNA bound to modern soil, points to sloughed off root-cap cells being the major source of plant DNA in sediments (del Pozo & Lam 1998; Willerslev et al. 2003). Additional sources may be pollen (mainly nuclear DNA) and leaf-litter. Furthermore, the action of microbial enzymes and pathogens can facilitate the release of DNA into the rhizosphere (Meier & Wackemagel 2003; Pote et al. 2005). Experimental evidence suggests that faeces, skin flakes and chitinious material such as hair, feathers and nails are major sources of DNA from vertebrates (Lydolph et al. 2005). In all cases the tissue may have disintegrated, releasing DNA to the surroundings, to the extent that no visible macrofossil traces are left behind. Interestingly, for both plant and animals, it has been shown that the 'dirt' DNA (mtDNA and cpDNA) is of regional origin and that long-distance dispersed genetic material seems to be insignificant (Haile et al. 2007; Willerslev et al. 2007).

Intriguingly, most ancient specimens contain only minor, if any, amplifiable endogenous DNA that additionally is highly degraded (Gilbert et al. 2005; Willerslev & Cooper 2005; Binladen et al. 2006). This holds for all ancient DNA sources including that of 'dirt'. Therefore, upper time limits exist on DNA survival that is highly dependent on the exact conditions of preservation (Willerslev et al. 2004b). This, coupled with the enormous amplification power of the Polymerase Chain Reaction (PCR) used to retrieve ancient DNA sequences, creates a huge risk of obtaining false positive results due to contamination with contemporary DNA (Hebsgaard et al. 2005). Thus, the authentication of ancient DNA results demands a heavy burden of proof (Willerslev & Cooper 2005).

An additional problem faced in 'dirt' DNA studies is the risk of DNA leaching between strata obscuring the temporal context of the data. Although there is strong evidence suggesting that free DNA, in sediments surviving degradation and metabolism by bacterial and fungal exonucleases, will quickly bind to clay, sand, humic substances and organomineral complexes in the sediment (Crecchio & Stotzky 1998), DNA leaching has proved a problem under certain settings. …